[{"data":1,"prerenderedAt":360},["ShallowReactive",2],{"project-43846964":3},{"id":4,"name":5,"line":6,"line_source":7,"title":8,"url":9,"domain":10,"maker":11,"posted_at":12,"year":13,"points":14,"comments":15,"rank_in_year":16,"shown_in_year":17,"top":18,"tier":19,"category":20,"subcategory":21,"format":22,"keywords":23,"reception":29,"sentiment":30,"stance_mix":31,"scored_comments":36,"aspects":37,"strip":48,"post_text":50,"what":51,"said":100,"maker_in_thread":215,"shape":224,"grid":227,"attempts":322,"same_week":323,"maker_shows":272,"maker_hidden":352,"reader_version":353,"card":354},"43846964","Promptrepo","Create your own finetuned AI model using Google Sheets","title","Show HN: Create your own finetuned AI model using Google Sheets","https:\u002F\u002Fpromptrepo.com\u002Ffinetune\u002F","promptrepo.com","QueensGambit",1746028416,2025,137,41,547,3837,15,"notable","ai","model fine-tuning platform","web_app",[24,25,26,27,28],"fine-tuning","llm fine-tuning","google sheets","no-code ai","openai fine-tuning","curious",-0.055,{"positive":32,"neutral":33,"mixed":34,"critical":35},4,16,2,7,29,{"idea":38,"pricing":40,"execution":43,"docs":44,"performance":45,"design":46,"privacy":47},{"positive":39,"critical":39},1,{"positive":41,"critical":42},0,3,{"positive":39,"critical":41},{"positive":41,"critical":41},{"positive":41,"critical":34},{"positive":34,"critical":41},{"positive":41,"critical":39},[42,42,41,42,42,49,42,39,41,39,41,41,42,42,39,42,42,39,42,39,49,41,49,49,42,42,39,41,41,42,42,41,41,42,41,41,42,42,39,42],5,"Hello HN,\n\nWe built Promptrepo to make finetuning accessible to product teams — not just ML engineers. Last week, OpenAI’s CPO shared how they use fine-tuning for everything from customer support to deep research, and called it the future for serious AI teams. Yet most teams I know still rely on prompting, because fine-tuning is too technical, while the people who have the training data (product managers and domain experts) are often non-technical. With Promptrepo, they can now:\n\n- Add training examples in Google Sheets\n\n- Click a button to train\n\n- Deploy and test instantly\n\n- Use OpenAI, Cla",{"problem":52,"audience":58,"pricing":66,"motivation":75,"time_to_build":80,"stack":81,"openness":98,"stage":99,"team_size":98},{"value":53,"source":54,"evidence":55},"Fine-tuning is too technical for product teams whose domain experts hold the training data","maker",[56],{"id":57},"post",{"value":59,"source":54,"evidence":60},"Product teams and non-technical domain experts",[61,62],{"id":57},{"id":63,"by":64,"text":65},"43855115","manidoraisamy","Our target customers are product teams like ours. I’m the technical founder and was building a feature to extract product prices from unstructured text. But my co-founder, who…",{"value":67,"source":54,"evidence":68},"Free trial; basic self-managed plan with bring-your-own OpenAI account; monthly and yearly plans, trial extendable up to 3 months",[69,72],{"id":70,"by":64,"text":71},"43848701","Thanks! Yes, that's a popular request from other developers as well. That's why our basic plan is self managed plan (Bring your own OpenAI account) -…",{"id":73,"by":11,"text":74},"43850413","I understand your concern. We’ve had cases where short-term users abused our product for phishing, which led us to remove the monthly plan initially. To address that without…",{"value":76,"source":54,"evidence":77},"Product teams used it internally for years and want non-technical staff to improve models without engineers",[78,79],{"id":57},{"id":63,"by":64,"text":65},null,[82,86,90,94],{"value":83,"source":54,"evidence":84},"OpenAI",[85],{"id":57},{"value":87,"source":54,"evidence":88},"Claude",[89],{"id":57},{"value":91,"source":54,"evidence":92},"Gemini",[93],{"id":57},{"value":95,"source":54,"evidence":96},"Llama",[97],{"id":57},"unknown","launched",{"praise":101,"objections":125,"requests":161,"questions":181,"alternatives":203},[102,118],{"value":103,"source":104,"evidence":105},"Google Sheets interface is simple and accessible","commenter",[106,110,114],{"id":107,"by":108,"text":109},"43853235","ygreif","I love integrating into spreadsheets. Super easy to use. Reminds me a bit of mailmerge. I don't have much experience with modern finetuning, but isn't it highly technical. How…",{"id":111,"by":112,"text":113},"43849136","ivape","This is awesome, I don't have a fine-tuning use case yet, but I can't imagine something being easier than a spreadsheet.",{"id":115,"by":116,"text":117},"43849800","rallyvite","Agree, not super familiar with this space but that you are using GS as the UI makes this super accessible to so many more people given familiarity and low intimidation factor.…",{"value":119,"source":104,"evidence":120},"Good UI idea for non-technical people giving model feedback",[121],{"id":122,"by":123,"text":124},"43848565","rgbrgb","Incredibly crowded space, but this is a great insight and UI... engineers probably have to integrate the model but we should empower non-technical customer-facing people to give…",[126,137,147,154],{"value":127,"source":104,"evidence":128},"Pricing feels tricky; annual plan push seen as a trap",[129,133],{"id":130,"by":131,"text":132},"43850250","ipaddr","I don't like the pricing. $38 a month headline much smaller billed annually with the price more than doubling. This is the trick your customers strategy and hope for no…",{"id":134,"by":135,"text":136},"43851661","jnovek","When I see a product — especially an early-stage product — trying to funnel me into a yearly subscription, the signal I get is that you think I’ll quit in less than a year which,…",{"value":138,"source":104,"evidence":139},"Storing training data in Google Sheets raises privacy concerns",[140,144],{"id":141,"by":142,"text":143},"43850552","littlestymaar","Why Google Sheet though? Why would you want your customers to give their training data to Google?!",{"id":145,"by":142,"text":146},"43854420","You are still giving your data away to Google, and it's one line of ToS change from allowing them to do it. And we know this Lind of changes happen all the time.",{"value":148,"source":104,"evidence":149},"Unclear who the target customer is and what the service does",[150],{"id":151,"by":152,"text":153},"43854592","raylad","Who is your target customer? From reading the site, it's not clear that someone who doesn't already know about fine tuning and want to do it would know what your service does or…",{"value":155,"source":104,"evidence":156},"Hard to prove fine-tuning beats prompting and RAG",[157],{"id":158,"by":159,"text":160},"43848635","Centigonal","Your biggest obstacle is proving fine-tuning is more effective than prompting, workflow design, RAG, etc during the initial pass. Most of my customers are still getting big…",[162,170,174],{"value":163,"source":104,"evidence":164},"Bring-your-own API key option with credits",[165,166],{"id":122,"by":123,"text":124},{"id":167,"by":168,"text":169},"43849871","heresjohnny","Neat! Wonder though if you should be even offering the BYO option as a separate lite package. As a dev I would not buy this and as a non-tech person I would be confused by your…",{"value":171,"source":104,"evidence":172},"Clearer examples and target use cases on the site",[173],{"id":151,"by":152,"text":153},{"value":175,"source":104,"evidence":176},"Automated evals, latency and cost comparisons across models",[177],{"id":178,"by":179,"text":180},"43848857","justanotheratom","FineTuning Ops platform requirement - compare evals, latency, cost across models.",[182,189,196],{"value":183,"source":104,"evidence":184},"How is this different from OpenAI projects?",[185],{"id":186,"by":187,"text":188},"43849819","labrador","How is this different than OpenAI projects?",{"value":190,"source":104,"evidence":191},"How is overfitting avoided?",[192],{"id":193,"by":194,"text":195},"43862223","sneha_tamal","How do you make sure the fine-tuning process stays effective without the risk of overfitting on specific data?",{"value":197,"source":104,"evidence":198},"Can fine-tuned models be used outside the platform?",[199],{"id":200,"by":201,"text":202},"43850765","mogili","I am not saying there isn't value for this, just saying what it does. I might actually use this if they don't tie the fine tuned models to only be used within their platform.",[204,211],{"value":205,"source":104,"evidence":206},"Kiln AI, a custom UI for fine-tuning, evals and synthetic data",[207],{"id":208,"by":209,"text":210},"43850097","scosman","What’s the thinking of spreadsheet first? Just making it super accessible for people who already have data? I’m building a UI for fine tuning (and evals, and synthetic data gen)…",{"value":212,"source":104,"evidence":213},"OpenAI projects in ChatGPT",[214],{"id":186,"by":187,"text":188},{"comments":49,"answered":34,"key":216},[217,218,221],{"id":73,"text":74},{"id":219,"text":220},"43852113","There's some truth to that. Early stage products often start out rough and part of the journey is finding early users who believe in the potential enough to stick around while we…",{"id":222,"text":223},"43852756","Yeah, Google Sheets is a surprisingly powerful interface for business teams, especially when they’re the ones curating the data. Curious to see how Promptrepo fits into your…",{"commenters":225,"max_depth":49,"maker_replies":49,"first_comment_mins":226,"lifespan_hours":36},22,103,[228,230,232,236,238,240,242,244,246,247,248,249,253,254,256,257,260,261,265,268,270,275,277,280,281,283,285,288,291,293,295,297,302,304,306,308,310,312,315,318,320],[158,41,42,229,159,41],"project\u002Fneutral\u002Fidea",[231,41,42,229,64,41],"43848851",[233,41,41,234,235,41],"43851163","commenter\u002Fneutral\u002Fnone","polskibus",[122,39,42,237,123,41],"project\u002Fmixed\u002Fpricing",[70,39,42,239,64,41],"project\u002Fneutral\u002Fpricing",[107,34,49,241,108,41],"project\u002Fpositive\u002Fexecution",[243,34,42,229,64,41],"43854985",[130,42,39,245,131,41],"project\u002Fcritical\u002Fpricing",[73,42,41,239,11,39],[134,42,39,245,135,41],[219,42,41,239,11,39],[250,42,41,251,252,41],"43854112","commenter\u002Fneutral\u002Fpricing","devjab",[208,32,42,229,209,41],[255,32,42,229,64,41],"43850170",[167,49,39,245,168,41],[151,258,42,259,152,41],6,"project\u002Fneutral\u002Fdocs",[63,258,42,229,64,41],[262,35,39,263,264,41],"43848419","project\u002Fcritical\u002Fperformance","nbbaier",[266,35,42,267,64,41],"43848499","project\u002Fneutral\u002Fperformance",[269,35,39,263,264,41],"43848759",[271,272,49,273,274,41],"43852692",8,"project\u002Fpositive\u002Fidea","ricktdotorg",[222,272,41,276,11,39],"project\u002Fneutral\u002Fnone",[111,278,49,279,112,41],9,"project\u002Fpositive\u002Fdesign",[115,278,49,279,116,41],[193,282,42,276,194,41],10,[284,282,42,267,64,41],"43862987",[141,286,39,287,142,41],11,"project\u002Fcritical\u002Fprivacy",[289,286,41,290,11,39],"43850610","project\u002Fneutral\u002Fprivacy",[145,286,41,292,142,41],"commenter\u002Fcritical\u002Fprivacy",[178,294,42,276,179,41],12,[296,294,42,276,64,41],"43848926",[298,299,41,300,301,41],"43851602",13,"other\u002Fneutral\u002Fnone","gitroom",[303,299,41,229,11,39],"43851709",[186,305,42,229,187,41],14,[307,305,41,234,64,41],"43849956",[309,305,41,234,187,41],"43850094",[311,305,42,229,64,41],"43850285",[313,18,42,314,201,41],"43849410","project\u002Fneutral\u002Fexecution",[316,18,39,317,64,41],"43849465","project\u002Fcritical\u002Fidea",[319,18,41,300,112,41],"43849684",[200,18,42,321,201,41],"project\u002Fmixed\u002Fnone",[],[324,329,334,339,344,348],{"id":325,"name":326,"points":327,"category":328},"43820228","PyXL",983,"hardware",{"id":330,"name":331,"points":332,"category":333},"43803148","RetrOS-32",575,"infrastructure",{"id":335,"name":336,"points":337,"category":338},"43816419","screenrecorder.me",462,"dev_tools",{"id":340,"name":341,"points":342,"category":343},"43835584","Beatsync",439,"creative",{"id":345,"name":346,"points":347,"category":333},"43778081","TacOS",320,{"id":349,"name":350,"points":351,"category":343},"43791992","Clever Coloring Book",316,false,"v2",{"id":355,"name":5,"line":6,"ai":352,"cat":20,"year":13,"points":14,"comments":15,"rec":29,"s":356,"top":18,"kw":357,"maker":11,"domain":10,"tier":19,"attempts":41,"strip":358,"url":9,"mix":359,"praise":103,"objection":127,"sub":21},43846964,-0.05,[24,25,26,27,28],"3303353101003313313150553310033003003313",[32,33,34,35],1794036673942]